Although the new Data Center Optimization Initiative (DCOI) strategy focuses on improving efficiencies in the largest centers, agencies cannot afford to ignore the seemingly countless small server facilities that support daily operations.
Beyond the requirements of DCOI, agencies have a vested interest in gaining visibility into the full scope of their compute resources, including both the small centers often found in remote offices and the numerous server rooms scattered around agencies.
These smaller facilities, usually classified as non-tiered data centers, do not offer the same potential for cost-savings found in full-scale data centers, but as the Government Accountability Office noted in a recent report, they still “pose security risks and management challenges and are an inefficient use of resources.”
The latest DCOI memo acknowledges this reality. While emphasizing the importance of optimizing full-fledged data center operations, the Office of Management and Budget notes that “agencies should consolidate any business applications hosted in non-tiered data centers and closets when cost-effective and as appropriate for accomplishing the agency mission and enhancing system security or information privacy.”
In our experience, the key to making these smaller compute centers more cost-effective and more secure is better operational intelligence.
In many cases, agencies cannot gauge the efficiency or security of these operations because they do not have an accurate inventory of the resources involved. Simply put, you cannot protect or report on what you can’t see. That’s what makes optimization or consolidation of these smaller centers cost-prohibitive: The work required to sort out what’s what eats into any savings that might be gained.
The lack of visibility can stymie consolidation efforts even when the case for consolidation appears to be clear cut. For example, as they shut down small facilities, agencies should look to consolidate server hardware and eliminate redundant software licenses. But that’s possible only if they have a clear accounting of their software licenses, where that software is running and the utilization rates — part of what we call operational intelligence.
Operational intelligence should incorporate the following functions:
- Live software asset inventory and utilization data. Data center administrators should be able to access real-time software utilization metrics from all endpoints and use that data to perform regular audits and to eliminate expensive unused or underused licenses.
- Detailed up-to-date utilization data from all environments. At a broader level, administrators should have the tools to produce an accurate inventory of every resource in the enterprise, from workstations and servers to virtual machines and cloud-based assets.
- Integration with existing management tools. Agencies should leverage the information described above to support on-going management of their environment. That includes integration with help desk operations, configuration management databases and asset management systems.
IT asset visibility and management traditionally involved arduous manual processes that tended to fall short of expectations. And those processes, being manual, often resulted in inaccurate data, leaving administrators to rely on a fair amount of guesswork when making critical decisions about how to adapt to changing performance requirements.
But today, there’s no need to fall back on guesswork. The IT infrastructure provides a wealth of data that agencies can use to get enterprise visibility at a granular level.
Clearly, such operational intelligence makes it much easier for agencies to comply with the reporting requirements of DCOI, as well as the MEGABYTE [Making Electronic Government Accountable By Yielding Tangible Efficiencies] Act, which requires CIOs to develop a plan for managing software licenses.
But the value of operational intelligence goes beyond compliance efforts. First, it helps agencies reduce wasted spend and so frees up money for innovation and modernization efforts. Also, it provides the foundation for automated infrastructure management, which improves overall enterprise performance.
Indeed, the goal of operational intelligence — like the goal of DCOI itself — is not to solve a technical problem, but rather a business one: Given the current budget constraints, how can agencies better leverage their existing resources to support mission objectives?
Operational intelligence supports both DCOI and its larger aim by providing agency leaders with the information they need to make good business decisions. Regardless of how DCOI requirements evolve in the future, the need for informed management decisions will be a constant.